Influence of liberation of sulphide minerals on flotation of sedimentary copper ore

Transcription

1 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ Influence of liberation of sulphide minerals on flotation of sedimentary copper ore Alicja Bakalarz 1,*, Magdalena Duchnowska 1, and Andrzej Luszczkiewicz 1 1 Wroclaw University of Science and Technology, Wybrzeze Wyspianskiego 27, -3 Wroclaw, Poland Abstract. Ore liberation is one of the most important parameter in mineral processing, especially in flotation. To separate the valuable components from gangue minerals, it is necessary to liberate one from others. It is achieve primarily through crushing and grinding. These stages are one of the most expensive of mineral production. It is important to determine the adequate mineral liberation which would result in huge savings in the overall cost of flotation plant. The aim of the paper was the analysis of the influence of milling time on the laboratory flotation of the copper ore from stratiform Polish deposit. Three different milling time of copper ore in laboratory ball mill was applied. The flotation results were presented as the recovery-recovery and grade-recovery upgrading curves. The liberation of sulphides and the particle size of sulphides in flotation product were analysed and compared. 1 Introduction The success of the separation process is related to the size and liberation of feed particles [1-3]. The definition of liberation varies from author to author [4]. The feed can consist of a single mineral called free particles, or it may consist of two or more minerals named locked particles. Locked particles can be binary, ternary, quaternary etc., as they consist of two, three, four or more minerals [5]. Liberation of particles is achieved mainly through crushing and grinding of the ore. These stages usually consume 3 of overall energy costs of mining operation, but it can rise even to for hard and/or finely intergrown ores. Achieving sufficient liberation can bring huge savings in costs of mineral production [6-8]. Flotation process strictly depends on the particle size. According to Pease et al [9], the finer particles tend to be more liberated, different than coarse particles. Nevertheless, the recovery is reduced for finer and coarser particles. Gaudin [5] as a first extensively described the liberation issue associated with comminution by two means: liberation by size reduction and liberation by detachment. According to the paper, the less abundant mineral is not fully liberated unless the particles are finer than the grain size. Moreover, the more abundant phase is always better liberated than the less abundant phase. The liberation of the ore is dependent on the mineralogy and texture [4,5,1]. Therefore, the upgrading results of flotation stream depend on the texture of the ore and the particle size to which the processing ore is crushed or milled. In Fig. 1 is shown, how particle composition defines the theoretical grade recovery curve [11]. It is important, that this type of curve do not take account of the process kinetics and mechanical entrainment [1]. According to Cropp et al [11] the losses in tailings are caused mainly by fine and liberated valuable minerals and locked minerals. The reasons of reducing the grade of concentrate are gangue composites, entrained and activated gangue, and deleterious element distribution in various size fractions. Fig. 1. A schematic of the theoretical grade recovery curve with typical particle images included (left) and the position of the actual grade/recovery relative to the theoretical potential (right); the area 1 the upgrading can be improved by operational changes; the area 2 grinding and classification is necessary to improve the upgrading results (based on Cropp et al [11]). Inadequate liberation is often the main reason of the poor performance. In the literature is a lot of examples, where process mineralogy surveys identified and resolved many problems in the process. The mineralogical and liberation analysis can result in addition of a regrind mill in the circuit or an application of the metallurgical operations in the upgrading stream [1]. The modern mineral liberation measurement techniques can be separated into two main types: two and three dimensional. Two dimensional technique is used in the automated scanning electron microscopy with energy * Corresponding author: alicja.bakalarz@pwr.edu.pl The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4. (

2 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ dispersive X-ray detectors (SEM/EDS), which popularity and application have been raising since s of the twentieth century. Nowadays automated image analysis instruments widely available are QEMSCAN, MLA or EPMA [1,12]. Three dimensional techniques relate to the non intrusive measurements of the whole ore particle (e.g. 3D tomography). This paper focuses on the influence of liberation of sulphide minerals in flotation of Polish copper ore from the Lubin-Glogow Basin (LGOM, SW Poland), the biggest copper deposit in Europe [13]. To prepare the flotation feed, different milling time of ore was applied. Mineralogical analysis of all upgrading products using QEMSCAN system were conducted. The influence of liberation of copper sulphide minerals in narrow particle size on the upgrading selectivity was analysed by plotting separation curves. 2 Experimental 2.1. Materials and methods A sample of sandstone copper ore collected from Rudna mining area was investigated. The chemical and mineral composition of flotation feed is presented in Table 1. The sample contains 1.72 of copper. The flotation feed contains 59 quartz, 21 carbonates and 1 clay minerals and micas. The content of all sulphides is about 2.8 (1.3 chalcocite,.8 bornite,.2 chalcopyrite and.3 pyrite). Additionally, the flotation feed contained.1 covellite,.5 galena and.4 sphalerite. The ore was dry-crushed in the jaw crusher to obtain particles fraction below 1 mm. A sample of g in the presence of cm 3 of industrial water was wet-ground Feed g in a laboratory ball mill. The grinding media (stainless balls) filled about of the mill capacity. Three different milling time of copper ore was applied:, and minutes. Table 1. Chemical and mineral composition of flotation feed. Element/Mineral Content, Cu 1.72 Chalcocite 1.35 Bornite.76 Chalcopyrite.21 Covellite.14 Pyrite.26 Galena.5 Sphalerite.4 Quartz Ca, Mg carbonates Clay minerals+micas 1. Others 5.3 The flotation methodology is presented in Fig. 2. All experiments were conducted in a Denver D12 laboratory flotation machine equipped with 2.5 and 1.5 dm 3 flotation cells. The air flow rate was regulated during each flotation test using a rotameter. Industrial mixture of sodium ethyl and isobutyl xanthate and sodium O,O-diethyl dithiophosphate in proportion 7:3 and the dose of g/mg was used as a collector. An aqueous solution of Nasfroth was utilized as a frother at a dosage of 2 g/t in all flotation tests. Both reagents were prepared directly before flotation test. After the experimental part, all the products were dried and weighted. A QEMSCAN system was used to determine mineralogy of feed samples and flotation products. The content of copper was determined by using a spectrometric method. 2,5 dm 3 stirrer speed: 1 rpm flotation time: 35' air: 12 dm 3 /h Tailings 1,5 dm 3 1,5 dm 3 1,5 dm 3 1,5 dm 3 stirrer speed: rpm flotation time: 1.5' air: 2 dm 3 /h stirrer speed: rpm flotation time: 2.5' air: 2 dm 3 /h stirrer speed: rpm flotation time: 6' air: below dm 3 /h stirrer speed: rpm flotation time: 1' air: below dm 3 /h 2,5 dm 3 stirrer speed: 12 rpm flotation time: 25' air: dm 3 /h P1 1,5 dm 3 stirrer speed: rpm flotation time: 2' P2 air: dm 3 /h C1 C2 C3 C4 ` P3 Fig. 2. The scheme of experimental methodology. 2.2 Results Upgrading results The results of flotation were evaluated by the mass balance of components and products of separation as well as by plotting recovery-recovery (the Fuerstenau upgrading curve, Fig. 3) and grade-recovery separation curves (Fig. 4). As it is presented in Figs. 3 and 4, the best and similar results of upgrading of copper sulphides were obtained after and minutes of milling. In these tests the best upgrading selectivity of copper sulphides was observed. It can be assumed that the milling time equals to is sufficient to obtain the highest upgrading selectivity of the examined copper ore. 2

3 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ tailing, r, Fig. 3. A recovery of copper sulphides in concentrate vs. cumulative recovery of remaining components in tailing obtained after, and minutes of milling. Grade of copper sulphides, β, Fig. 4. Copper sulphide grade vs. recovery obtained after, and minutes of milling Analysis of liberation results The comparison of the grain composition and liberation of sulphides in the flotation feed is presented in Fig. 5 and 6. Cumulative yield of size particle in flotation feed, 2 Recovery of copper sulphides in concentrate,, copper sulphides minutes copper sulphides minutes Recovery of copper sulphides in concentrate,, minutes d max, mm Fig. 5. Particle size distribution in flotation feeds. As expected, the particle size decreases with increasing milling time (Fig. 5). After, and minutes of milling, P parameter equals to 52 um, 46 um and 38 um, respectively. In table 2 the liberation of sulphides in the feed after milling is presented. The results of liberation was splited on four ranges: 2 (locked particles), 2 and (impregnations) and (free particles). It is important to mark that these data were calculated for sulphides of copper (chalcocite, chalcopyrite, bornite, covellite and tennantite), zinc (sphalerite), lead (galena) and iron (pyrite). As expected, the liberation of sulphides becomes better with the milling time of the feed decreasing (Table 2). It is noticeable in each grain fraction. The highest of fully liberated particles are observed for particles size in the range of 2 um. It is important to note that quite small contents of impregnations are observed in each fraction in feed. The relationship between liberation of sulphides and milling time is presented in Fig. 6. Similar results of liberation in flotation feeds after milling in and minutes were obtained. A significant differences are observed in liberation of sulphides in feed after minutes and or minutes. It indicates that milling time equals to is sufficient to obtain the maximum of liberation of sulphide minerals in examined copper ore. However, it is important to notice, that this observation concerns in the treatment the ore in a wide size distribution. The liberation of sulphides in narrow grain fractions in flotation feed were analysed (Figs. 7-8). Table 2. Liberation of sulphides in flotation feed. Particle size, um > Liberation, Milling time of the feed, minutes

4 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ Cumulative yield of sulphides in flotation feed, Fig. 6. Liberation of sulphide minerals in flotation feeds. Cumulative yield of sulphides in flotation feed, Fig. 7. Liberation of sulphide particles in the size range of 2 um in flotation feed. Cumulative yield of sulphides in flotation feed, 2 minutes Liberation of sulphides minutes Liberation of sulphides minutes > um Liberation of sulphides Fig. 8. Liberation of sulphide particles in the size range of > um in flotation feed. As it has been shown, significantly worse liberation after of milling of each grain fraction was observed. It is presented in Fig. 7, that the comminution of sulphides to the size minus 2 um does not provide the good liberation of the fines. Very similar results were observed for sulphide particles in the range 2 to um. Similar liberation for sulphides in the size range of 71 um after and minutes of milling was obtained. Probably, the reason of weak liberation of fines is very small dimensions of sulphides, which initially appears in the ore. It can be supposed that the grinding of narrow grain fractions separately can result in better liberation of sulphides, also for fines. Noticeable differences between and minutes of milling are observed for particle size ranging from 71 to um and > um. For the grain fraction > um, the liberation of sulphides increases with the milling time (Fig. 8). The influence of liberation of sulphides was evaluated by plotting grade-recovery and recovery-recovery (α-nonsensitive) upgrading curves. The analysis was conducted for narrow particle size fractions and the liberation degree of sulphides. Only the part of figures is shown in the paper. The curves are presented in Figs Recovery of remaining components in tailing, Fig. 9. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in size range of 2 um obtained after, and minutes of milling. Grade of particle size in concentrate, minutes Cu sulphides Cu sulphides minutes Fig. 1. Copper sulphides grade vs. recovery in size range of 2 um obtained after, and minutes of milling. As it is presented in Figs. 9-1, the lowest upgrading selectivity and grade of the finest particles in concentrate is observed in flotation test in which of milling was applied. Similar performance was obtained for the flotation experiments after longer milling time ( and minutes). Comparable results were obtained for sulphides in the size range of 2 um. Since the each flotation feed contains the highest content of grain fraction between and um size (Fig. 3), lower upgrading selectivity of these sulphides size can result in overall poorer upgrading of copper. 4

5 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ Recovery of remaining components in tailing, -71 um Cu sulphides minutes Fig. 11. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in size range of 71 um obtained after, and minutes of milling. For the particles greater than um, the differences between three curves become more significant with extending of the milling time (Fig. 11). Similar results for the particles size ranging from 71 to um were obtained. The best upgrading selectivity for copper sulphides was observed after milling time equals to minutes. A bit worse flotation efficiency after minutes of milling was obtained. The lowest selectivity is observed for the shortest milling time. For the sulphide grains greater than um the observation has changed (Fig. 12). It is quite interesting that even for coarse-grain fraction (+ um) the best upgrading selectivity is obtained for the flotation feed after minutes of milling. tailing, > um Cu sulphides minutes Fig. 12. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in grain fraction greater than um obtained after, and minutes of milling. In Figs the flotation results of free (liberated) sulphides particles is presented. As it can be seen in Fig. 13, the highest and similar upgrading selectivity of sulphides is observed for tests with milling time of and minutes. The lower selectivity was obtained after feed milling in. Similar results were obtained for the particle size in the range of 2 71 um. The comparable dependency is presented in Fig. 14 as grade-recovery curves. The quality of concentrate differs significantly for all tests. tailing, minutes Fig. 13. A recovery of sulphides vs. recovery of remaining components in tailing in size range of 2 um liberated in (free particles) obtained after, and minutes of milling. Grade of particle size in concentrate, liberation: - Fig. 14. Sulphides recovery vs. grade in size range of 2 um liberated in (free particles) obtained after, and minutes of milling. tailing, liberation: - minutes 71- um libertaion:- minutes Fig. 15. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in size range of 71 um liberated in (free particles) obtained after, and minutes of milling. 5

6 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ More significant differences are observed with increasing of particle size (Fig. 15). The upgrading selectivity of free particles flotation decrease in order:, and minutes of milling. The influence of the particle size on the upgrading selectivity is presented in Fig. 16. It can be seen that after milling time equals to minutes, the flotation efficiency decreases while the size of the sulphides becomes smaller. The obtained results confirms data from the literature. Fines, especially in the size range of 2/ um, float worse than coarse sulphides, even fully liberated. Similar dependences for the rest of flotation experiments were obtained. One of the reason of poor floatability of fully liberated fines can be too less dosage of collector. It is confirmed that fines because of a larger particle surface area demand higher doses of collector than coarse. This issue can be solved by separately upgrading fines from coarse. tailing, um -71 um 71- um > um Fig. 16. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing for sulphides liberated in (free particles) obtained after of milling. tailing, 95 liberation: - liberation: - minutes Fig. 17. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in particle size of 2 um liberated in (impregnation) obtained after, and minutes of milling. Next stage of the research was to study the flotation of sulphides in impregnations (Figs ). It was observed that there is no significant differences between two ranges of liberation: 2 and. The results of the sulphides liberated in flotation for particle size ranging from of to 2 um and 2 to um were presented. As it can be seen that for the particles size in the range of 2 um the best upgrading selectivity was observed after and minutes of milling. Insignificantly worse efficiency was obtained for the test with the feed grinded in. Appreciably differences between the experiments are observed for particles in impregnations and greater than 2 um (Fig. 18). Similar results for the rest of results, not presented in the paper, were obtained. The poor upgrading selectivity was observed for the flotation test where the feed was grinded in. tailing, minutes Fig. 18. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in size range of 2 um liberated in (impregnation) obtained after, and minutes of milling. tailing, 2- um liberation: - liberation: -2 minutes Fig. 19. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing in particle size in the range of 2 um liberated in 2 (locked particles) obtained after, and minutes of milling. In Figs the flotation results of locked sulphides particles (liberation: 2) are presented. As it can be seen, the best upgrading selectivity of fines was obtained in the test with the feed grinded in minutes (Fig. 19). Similar results of flotation were observed for particle size in the range of 2 um. For the coarser particles (+ um) the differences between the plots courses become greater (Fig. 2). Despite the higher content of the locked 6

7 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ particles (+ um) in the flotation feed after of milling time, better upgrading selectivity of these particles was observed. The influence of the particle size on the upgrading selectivity for the test with milling time equals to minutes is presented in Fig. 21. It can be seen that the flotation efficiency increases while the size of the sulphides decreasing. These results are contrary to date obtained for fully liberated sulphides (Fig. 16). Grade of particle size in concentrate, um liberation: -2 minutes. 2 Fig. 2. Sulphides recovery vs. grade in particle size in range of 71 um liberated in 2 (locked particles) obtained after, and minutes of milling. tailing, liberation: um -71 um 71- um > um Recovery of particle size in concentrate Fig. 21. A recovery of copper sulphides in concentrate vs. recovery of remaining components in tailing for sulphides liberated in 2 (locked particles) obtained after of milling Summary and conclusions The influence of liberation and particle size of sulphide minerals on the laboratory flotation of sedimentary copper ore (Rudna mining area) is presented in the paper. The series of laboratory flotation experiments were performed. All flotation tests were conducted in the presence of collector and frother. Three different milling time of flotation feed were applied:, and minutes. Mineralogical analysis of all upgrading products using QEMSCAN system were conducted and analysed. After, and minutes of milling of the particles mass passing 52 um, 46 um and 38 um, respectively. In each flotation feed, the high contents of small particles ( um) were observed. It was shown that the milling time of it is enough to obtained the maximum of liberation of sulphide minerals. After minutes of milling the upgrading results were very similar. The milling time equals to is too short for the studied copper ore. Extremely similar observations were notice when it comes to liberation analysis. The comparable data of sulphides liberation for flotation feeds grinded by and minutes were obtained. For all flotation tests, similar results of sulphides liberation were observed for particle size in the range of um. For the greater particles more noticeable differences between sulphides minerals in flotation feeds were obtained. Liberation analysis confirmed upgrading selectivity of sulphides minerals in narrow fractions. The flotation efficiency was close similar for all particles in the size range of um in each flotation test. More noticeable differences between all tests were observed for coarser particles. Moreover, even for coarse particles (+ um) the best upgrading selectivity is obtained for the flotation feed after minutes of milling. Regardless of particle size and liberation degree, the best upgrading selectivity was obtained for experiments in which the milling time equals to or minutes was applied. Moreover, different dependence was observed for fully liberated and locked sulphides in analysed size fractions. With the bigger particle size, the higher upgrading selectivity of fully liberated sulphide minerals was observed. Quite different results was obtained for locked sulphide particles. The flotation efficiency increases with decrease of size of the sulphide minerals. These observations were noted for each experiment. It is confirmed that the different flotation needs of fine and coarse particles are essential. A treatment an ore in a wide size distribution can result in poorer upgrading results [14]. One of the reason is i.e. different requirements of collector dosages for fine and coarse particles. Fines because of a larger particle surface area demand higher doses of collector than coarse. Probably, the dosage of collector which was applied in all examined experiments was insufficient for fines while enough for coarser particles. Moreover, flotation kinetics of fines is slower. Good flotation of small but locked sulphides could be the effect of hydrodynamics and low weight of these particles. It can result in mechanical entrainment. Additionally, the flotation of that kind of particles can result in poorer quality of final concentrate. Concluding, with extending the milling time of copper sulphide ore, the liberation of valuable components becomes better, independently on particle size. In spite of poorer flotation of fully liberated and fine sulphides, deep grinding of sedimentary copper ore is essential to receive the greatest upgrading selectivity. Because of differences in flotation requirements for fine and coarser minerals, it seems to be justified to classification the feed into two or even more streams and processing in a narrow size distribution. The grinding of 7

8 E3S Web of Conferences 18, 125 (217 ) DOI:1.151/e3sconf/ narrow size fraction separately and then float should result in better liberation of sulphides of each particle size, better overall upgrading efficiency and lower power consumption. This issue requires further investigations. The research presented in the paper is the part of project titled Development of highly effective technology of Polish copper ore beneficiation. This project is carried out by the consortium with the Wroclaw University of Science and Technology as leader, agreements No.: NCBR: CuBR/I/7/NCBR/215 and KGHM Polska Miedz SA: KGHM-BZ-U References [1] W.J. Trahar, Int. J. Miner. Process. 8, (1981) [2] T. Leißner, T. Mütze, K. Bachmann, S. Rode, J. Gutzmer, U.A. Peuker, Miner. Eng. 53, (213) [3] J. Zhang, N. Subasinghe, Miner. Eng. 49, (213) [4] M.A. Berube, J.C. Marchand, Int. J. Miner. Process. 13, (1984) [5] A.M. Gaudin, Principles of Mineral Dressing (McGraw-Hill, New York. Henley, 1983) [6] R. Tomanec, J. Milovanovic, Physicochem. Probl. Mi. 28, (1994) [7] T.G. Vizcarra, E.M. Wightman, N.W. Johnson, E.V. Manlapig, Miner. Eng. 23, (21) [8] N. Djordjevic, Miner. Eng. 64, (214) [9] J.D. Pease, D.C. Curry, M.F. Young, Miner. Eng. 19, (26) [1] D. Bradshaw, Proceedings of IMPC, Santiago, Chile, October 214 (214) [11] A.F. Cropp, W.R. Goodall, D.J. Bradshaw, AusImm GeoMet 213 Proceedings, 279 (213) [12] N.O. Lotter, Miner. Eng. 24, (211) [13] C. Bachowski, J. Kudelko, H. Wirth, Biul. Państw. Inst. Geol., 423, (27) [14] J.D. Pease, M.F. Young, D.Curry, N.W. Johnson, MetPlant 24 AusIMM Proceedings (24) 8